The burden of pediatric HIV-1 infection globally is highest in sub-Saharan Africa with well over two million children infected currently. Recently, increased access to highly active antiretroviral therapy (HAART) promises hope of improved survival, but most HIV-1 infected children in sub-Saharan Africa initiate treatment at advanced disease stage and greater level of immunosuppression in contrast to children in Europe or the US. In addition, first line HAART regimens in Africa typically have a non-nucleoside reverse transcriptase inhibitor (NNRTI) backbone and do not include ritonavir-boosted protease inhibitors (Pis), which are considered the most potent antiretrovirals available. Failure to adequately suppress HIV-1 can rapidly lead to emergence of drug-resistance and negatively impact long-term response to HAART. The overall hypothesis of this proposal is that clinical disease stage, level of immunosuppression at presentation, and choice of initial HAART regimen will impact long-term response and selection of drug resistant virus. For this GRIP application, we propose to utilize and extend a recently accrued pediatric HIV-1 treatment cohort in Nairobi to determine the pattern and correlates of long-term response to HAART and to determine levels of resistance to first line regimens in HIV-1 infected children failing therapy. Our specific aims are: Aim 1: Define the pattern and predictors of clinical and virologic response to HAART during 5 years of follow-up in HIV-1 infected Kenyan children. Potential predictors include baseline CD4% and viral load, clinical disease severity, adherence, and choice of non-nucleoside reverse transcriptase inhibitor (NNRTI). Aim 2: Define the frequency of mutations associated with resistance to the 1st line NNRTI-based triple therapy being introduced in Kenya and the impact of such mutations on response to therapy in HIV-1 infected children. This study will provide important information on effectiveness of HAART regimens being scaled up in resource-poor countries and help programs better plan for changes of drug regimens following treatment failure. [unreadable] [unreadable] [unreadable]